Epilepsy & Psychiatric Comorbidities
Psychiatric comorbidities are among the most prevalent and impactful conditions in people with epilepsy, affecting quality of life, treatment adherence, seizure control, and mortality. Depression and anxiety each occur in 25–40% of people with epilepsy — rates 2–5 times higher than in the general population. The relationship between epilepsy and psychiatric disorders is bidirectional: not only does epilepsy increase the risk of developing psychiatric conditions, but premorbid depression and anxiety increase the risk of subsequently developing epilepsy. This bidirectional model, supported by both epidemiologic data and animal models, suggests shared underlying neurobiological mechanisms involving serotonergic, noradrenergic, and GABAergic systems. Clinicians must systematically screen for psychiatric comorbidities, understand the psychotropic profiles of antiseizure medications (ASMs), and approach treatment without fear of lowering the seizure threshold.
Bottom Line
- Depression: Prevalence 25–50% in epilepsy; NDDI-E (score >13) is the most validated screening tool; SSRIs (citalopram, escitalopram, sertraline) are first-line treatment with minimal interaction risk
- Anxiety: Prevalence ~40%; brEASI and GAD-7 validated in epilepsy; SSRIs are first-line; benzodiazepines are appropriate short-term but carry dependency risk
- Bidirectional relationship: Premorbid depression and stress increase epilepsy risk through serotonergic/noradrenergic depletion and enhanced kindling susceptibility
- ASM psychotropic effects: Levetiracetam (irritability, depression), perampanel (aggression), topiramate (depression, cognitive impairment) can worsen psychiatric symptoms; lamotrigine and valproate have mood-stabilizing properties
- Antidepressant safety: SSRIs and SNRIs at therapeutic doses do not meaningfully lower the seizure threshold; bupropion carries dose-dependent risk (especially >450 mg/day); clomipramine and clozapine have clinically relevant seizure risk
- Psychosis of epilepsy: Postictal psychosis is most common (60%); forced normalization (alternative psychosis) occurs with seizure freedom; avoid clozapine
- Suicidality: People with epilepsy have 3–5× increased suicide risk; the 2008 FDA warning on ASMs and suicidality has been questioned by subsequent meta-analyses showing no individual ASM consistently increases risk
Bidirectional Relationship Between Epilepsy & Psychiatric Disorders
The traditional view of psychiatric comorbidities as secondary consequences of epilepsy has been replaced by a bidirectional model. Population-based studies demonstrate that a history of depression increases the risk of developing epilepsy by 2–7-fold, and vice versa. In animal models, early postnatal stress enhances susceptibility to limbic epileptogenesis, and experimentally induced serotonin and norepinephrine depletion lowers the seizure threshold and facilitates kindling. These shared neurobiologic substrates suggest that epilepsy and depression/anxiety may be different manifestations of a common underlying brain dysfunction rather than a simple cause-and-effect relationship.
Five mechanisms of association between epilepsy and comorbidities have been proposed: (1) artifactual/chance association, (2) epilepsy causing the comorbidity (resultant), (3) the comorbidity causing epilepsy (causative), (4) shared risk factors, and (5) true bidirectional association. For cerebrovascular disease and epilepsy, bidirectionality is well-supported by clinical data. For depression and anxiety, the evidence from animal models is compelling, though human clinical evidence is still accumulating.
Depression in Epilepsy
Epidemiology & Impact
Depression is the most common psychiatric comorbidity in epilepsy, with prevalence estimates ranging from 25% to 50% depending on the population studied. The rate is highest in patients with drug-resistant temporal lobe epilepsy (approaching 50–60%) and in epilepsy monitoring unit populations. Depression in epilepsy is associated with decreased quality of life (accounting for more variance than seizure frequency), increased health care utilization, poorer medication adherence, increased risk of ASM intolerance and treatment failure, greater risk of SUDEP, and increased mortality. The presence of comorbid depression may be the single most important determinant of quality of life in people with epilepsy.
Screening Tools
| Instrument | Target Condition | Format | Positive Screen | Epilepsy-Specific? |
|---|---|---|---|---|
| NDDI-E | Major depressive disorder; suicidality | 6-item, 4-level Likert scale | Score >13 | Yes — designed to distinguish depression from ASM side effects and epilepsy-related cognitive effects |
| PHQ-9 | Major depressive disorder | 9-item, 4-level Likert scale | Score ≥10 | No — validated in epilepsy population |
| BDI-II | Major depressive disorder | 21 descriptive statements, 4-level Likert | Score >11 | No — validated in epilepsy population |
| HADS | Depression and anxiety | 14-item self-report | Score >7 (per subscale) | No — validated in epilepsy population |
| brEASI | Anxiety disorders | 8-item, 4-level Likert scale | Score ≥7 | Yes — derived from 18-item EASI |
| GAD-7 | Generalized anxiety disorder | 7-item, 4-level Likert scale | Score >6 | No — validated in epilepsy population |
Practical Screening Approach
- The ILAE recommends asking at least one mental health question at every visit
- The NDDI-E is the most efficient and practical tool for a busy epilepsy clinic — validated in 13 languages, takes <3 minutes, and screens for both depression and suicidality
- A positive screen is not diagnostic — it identifies patients who need referral for comprehensive psychiatric evaluation
- The National Academy of Medicine recommends routine use of validated screening instruments for mental health in people with epilepsy
- Screen at diagnosis, medication changes, seizure recurrence, and at least annually
Treatment of Depression in Epilepsy
The ILAE published clinical practice recommendations for the treatment of depression in adults with epilepsy in 2022. Key recommendations include:
- Mild depression: Psychological treatments (CBT, mindfulness-based approaches) are first-line (Level B — probably effective)
- Moderate to severe depression: SSRIs are first-line pharmacotherapy (Level B recommendation)
- SSRI preference: Citalopram, escitalopram, and sertraline are preferred due to minimal CYP450 interactions with ASMs
- Second-line: Venlafaxine (SNRI) is an appropriate second-line option (Level C — possibly effective)
- Combination therapy: Pharmacotherapy plus psychotherapy may be more effective than either alone
Anxiety Disorders in Epilepsy
Approximately 40% of people with epilepsy report anxiety symptoms, with generalized anxiety disorder, panic disorder, and social phobia being the most common. Anticipatory anxiety — the persistent fear of having a seizure in public, at work, or while driving — is highly prevalent and contributes significantly to social avoidance, agoraphobia, and functional impairment. This seizure-specific anxiety may not meet criteria for any standard DSM-5 anxiety disorder but causes substantial disability.
Temporal Relationship of Anxiety to Seizures
Anxiety in epilepsy has a complex temporal relationship to seizures that has important diagnostic and therapeutic implications:
| Temporal Pattern | Timing | Clinical Features | Management |
|---|---|---|---|
| Ictal anxiety | During seizure (aura) | Sudden, intense fear or panic lasting seconds to minutes; often stereotyped; common in mesial temporal lobe epilepsy; associated with amygdala activation | Optimize ASM for seizure control; may be misdiagnosed as panic disorder |
| Postictal anxiety | Minutes to hours after seizure | Generalized anxiety, agitation, or panic following a seizure; may last hours to days; often accompanies postictal depression | Self-limited; short-acting benzodiazepine if severe; reassurance |
| Interictal anxiety | Chronic, between seizures | GAD, social phobia, panic disorder; anticipatory anxiety about seizures; may fluctuate with seizure frequency | SSRI first-line; CBT; short-term benzodiazepine for acute episodes |
| Preictal anxiety | Hours before seizure | Prodromal anxiety or irritability preceding a seizure by hours; some patients can predict seizures based on this prodrome | Optimize ASM; rescue benzodiazepine; self-management strategies |
No specific evidence-based guidelines exist for treating anxiety in epilepsy; SSRIs remain reasonable first-line agents, and short-term benzodiazepine use is appropriate for acute anxiety but carries risks of dependence, cognitive impairment, and potential seizure exacerbation upon withdrawal. Gabapentin and pregabalin have anxiolytic properties and may serve a dual role as both ASM and anxiolytic in selected patients, though their efficacy as anxiolytics in epilepsy populations has not been specifically studied.
Psychosis of Epilepsy
Up to 6% of people with epilepsy have a history of psychosis. Psychotic episodes in epilepsy are categorized temporally in relation to seizures:
| Type | Proportion | Clinical Features | Pathophysiology | Treatment |
|---|---|---|---|---|
| Postictal psychosis | ~60% | Occurs after cluster of bilateral convulsive seizures; lucid interval of 1–3 days; affect-laden with paranoid/persecutory delusions; short duration (days to weeks); negative symptoms uncommon | Post-seizure rebound excitability; dopaminergic dysregulation; bilateral seizure spread | Low-dose atypical antipsychotic (olanzapine, risperidone) for short course; benzodiazepines; self-limited |
| Interictal psychosis | ~30% | Clinically can be indistinguishable from schizophrenia; may develop after years of poorly controlled epilepsy; chronic course possible | Chronic epileptic network disruption; hippocampal damage; limbic dysfunction | Olanzapine or risperidone first-line; quetiapine (interacts with enzyme-inducing ASMs); avoid clozapine |
| Forced normalization / Alternative psychosis | ~5–10% | Paradoxical development of psychotic symptoms when seizures are controlled and EEG normalizes; may occur after epilepsy surgery or ASM optimization | Proposed: seizure suppression unmasks latent psychotic vulnerability; neurochemical rebalancing | May require reducing ASM to allow some seizure recurrence; low-dose antipsychotic; psychiatric consultation |
| Ictal psychosis | Rare | Part of seizure semiology (usually complex partial status); hallucinations, paranoia during seizure; resolves with seizure termination | Direct effect of epileptic discharge, usually temporal lobe origin | Treat the seizure (benzodiazepine, ASM optimization) |
Antipsychotic Use in Epilepsy
- Clozapine: Contraindicated if clinically feasible — dose-dependent seizure risk of 3–5% (up to 10% at doses >600 mg/day); potent reduction of seizure threshold
- Chlorpromazine: Significant seizure risk, especially at high doses; avoid
- Olanzapine and risperidone: Preferred first-line antipsychotics in epilepsy; minimal seizure threshold effects
- Quetiapine: Acceptable alternative but interacts with enzyme-inducing ASMs (CYP3A4 substrate)
- Aripiprazole: Low seizure risk; reasonable alternative
- All antipsychotics carry theoretical seizure risk at supratherapeutic doses or with rapid titration
ADHD & Epilepsy
ADHD occurs in 20–40% of children with epilepsy and persists into adulthood in a substantial proportion. The relationship is likely bidirectional, with shared risk factors including genetic variants affecting neurodevelopment and neurotransmitter function. ASMs can exacerbate ADHD symptoms: phenobarbital and benzodiazepines cause inattention, while levetiracetam may cause irritability mimicking ADHD-related behavioral problems. Conversely, stimulant medications (methylphenidate, amphetamines) at therapeutic doses do not significantly increase seizure risk and are considered safe in patients with well-controlled epilepsy. Atomoxetine is an alternative nonstimulant option with no meaningful seizure threshold effects. The concern that stimulants lower the seizure threshold has not been borne out by clinical studies.
ASMs & Psychiatric Effects
| ASM | Beneficial Psychotropic Effects | Harmful Psychotropic Effects | Clinical Considerations |
|---|---|---|---|
| Lamotrigine | Reduced depression, mood stabilization, decreased mania | Insomnia; irritability (usually children with intellectual disability) | FDA-approved for bipolar maintenance; preferred ASM in patients with comorbid depression |
| Valproate | Decreased agitation, aggression, irritability, mania; mood stabilization | Depression (less common) | FDA-approved for bipolar mania; avoid in women of childbearing potential due to teratogenicity |
| Carbamazepine | Decreased aggression, mania; mood stabilization | Irritability, impaired attention | Historically used for mood disorders; extensive drug interactions limit utility |
| Levetiracetam | None established | Anxiety, depression, irritability; more common in children and patients with preexisting psychiatric history | Behavioral effects reported in 5–15% of patients; may require dose reduction or switch; pyridoxine (vitamin B6) may help irritability |
| Brivaracetam | None established | Similar to LEV but possibly lower rates of behavioral effects | May be better tolerated than levetiracetam in patients with psychiatric vulnerability |
| Perampanel | None established | Irritability, aggression (dose-dependent), depression, hostility | Psychiatric effects reported in 8–20% of patients; more common at doses ≥8 mg/day |
| Topiramate | Reduced binge eating, mania; mood stabilization | Depression, impaired cognition (word-finding, memory), attention difficulties, irritability | Cognitive and mood effects are dose-dependent; significant impact on quality of life |
| Phenobarbital | Anxiolytic, sedative | Aggression, impaired cognition and attention, depression, irritability, decreased libido | Most significant psychiatric adverse effect profile among older ASMs |
| Gabapentin/Pregabalin | Anxiolytic, sedative, mood stabilizing | Irritability, agitation (usually children with intellectual disability) | Gabapentin is FDA-approved for generalized anxiety (in some countries); useful for comorbid neuropathic pain |
| Zonisamide | Decreased mania | Aggression, emotional lability, irritability, depression | Psychiatric effects less common than topiramate but still notable |
Antidepressant Safety in Epilepsy
A longstanding and largely unjustified concern among neurologists has been that antidepressants lower the seizure threshold. This fear has contributed to significant undertreatment of depression in epilepsy. Current evidence strongly supports that SSRIs, SNRIs, and most antidepressants at therapeutic doses do not meaningfully increase seizure risk. In fact, some evidence suggests that SSRIs may have anticonvulsant properties through serotonergic mechanisms.
| Antidepressant Class | Seizure Risk | Specific Agents | Notes |
|---|---|---|---|
| SSRIs | Very low (may be anticonvulsant) | Citalopram, escitalopram, sertraline (preferred in epilepsy); fluoxetine, paroxetine | First-line for depression and anxiety in epilepsy; citalopram, escitalopram, sertraline minimize CYP450 interactions |
| SNRIs | Very low | Venlafaxine (ILAE second-line), duloxetine | Appropriate alternative; venlafaxine has best evidence after SSRIs |
| Bupropion | Dose-dependent; elevated at >450 mg/day | Bupropion IR, SR, XL | Immediate-release formulation has highest seizure risk; extended-release at ≤300 mg/day may be acceptable; use with caution |
| Tricyclic antidepressants | Low at therapeutic doses; high in overdose | Amitriptyline, nortriptyline, desipramine | Seizure risk primarily in overdose; anticholinergic effects problematic |
| Clomipramine | Elevated (∼1–2% at therapeutic doses) | Clomipramine | Highest seizure risk among tricyclics; avoid in epilepsy |
| Mirtazapine | Low | Mirtazapine | Sedating; may benefit insomnia; weight gain |
Suicidality & the FDA Warning
In 2008, the FDA issued a class-wide warning that all ASMs may increase the risk of suicidal ideation, suicide attempts, and completed suicide. This warning was based on a meta-analysis of 199 randomized controlled trials involving 11 ASMs, which found an odds ratio of 1.8 for suicidal ideation/behavior with ASMs compared to placebo. However, this warning has been extensively criticized and substantially undermined by subsequent evidence:
- The original meta-analysis was driven primarily by topiramate trials in non-epilepsy indications (migraine, weight loss) and may not be generalizable to epilepsy populations
- A 2021 meta-analysis by Klein et al. in JAMA Neurology examining 10 newer ASMs found no evidence of increased suicidality for any individual drug studied
- Longitudinal studies demonstrate that peak suicidality occurs in the period shortly before ASM prescription (ie, when seizures are uncontrolled) rather than after starting treatment
- The elevated baseline suicide risk in epilepsy (3–5-fold higher than the general population) is driven primarily by the disease itself, psychiatric comorbidities, and psychosocial burden — not by ASM treatment
- Uncontrolled seizures and failure to treat psychiatric comorbidities pose greater risk than ASM-associated suicidality
Managing Suicidality Risk in Epilepsy
- Screen for suicidality at every visit using the NDDI-E (validated for suicidality in epilepsy; item 4 specifically addresses thoughts of death)
- Directly ask about suicidal ideation — this does not increase risk and is considered standard of care
- Optimize seizure control — uncontrolled seizures are the greatest driver of psychiatric morbidity
- Treat comorbid depression and anxiety aggressively with SSRIs and/or psychotherapy
- Monitor patients initiating ASMs known to cause psychiatric effects (levetiracetam, perampanel, topiramate, phenobarbital) within the first 2–4 weeks
- Refer to psychiatry for patients with active suicidal ideation, prior suicide attempts, or treatment-resistant depression
- Do not withhold effective ASM therapy based solely on the 2008 FDA warning — the risk of uncontrolled seizures outweighs the theoretical ASM-related suicidality risk
Psychogenic Nonepileptic Events (PNES)
Psychogenic nonepileptic events (PNES, formerly "pseudoseizures") are a critical differential diagnostic consideration in any epilepsy clinic. PNES are paroxysmal episodes of altered movement, sensation, or experience that resemble epileptic seizures but are not associated with epileptiform EEG correlates. They are classified as a functional neurological disorder (conversion disorder) and represent the second most common diagnosis made in epilepsy monitoring units, accounting for 20–40% of admissions.
| Feature | Epileptic Seizures | PNES |
|---|---|---|
| Duration | Usually <2 minutes (GTC); variable for focal | Often prolonged (>2 minutes, may last 10–60 minutes) |
| Motor pattern | Stereotyped; tonic phase → clonic phase with declining frequency | Waxing/waning intensity; asynchronous movements; pelvic thrusting; side-to-side head movements |
| Eye closure | Eyes usually open during GTC seizure | Eyes frequently closed and may resist opening |
| Postictal confusion | Typical after GTC; gradual recovery | Variable; may have rapid return to baseline; crying common |
| Tongue biting | Lateral tongue biting (specific for epilepsy) | Tip of tongue biting (less specific) |
| Urinary incontinence | Common in GTC | Can occur (less specific than previously thought) |
| Prolactin elevation | Elevated 15–20 minutes after GTC (sensitivity ~60%) | Normal |
| EEG correlate | Epileptiform activity during event | No epileptiform activity; normal background or muscle artifact |
| Psychiatric history | Variable | PTSD, trauma history, dissociative symptoms common (60–80%) |
Key Principles in PNES Management
- Diagnosis: Video-EEG monitoring with capture of a typical event is the gold standard; semiologic features alone are insufficient for definitive diagnosis
- Coexistence: 10–30% of patients with PNES also have epilepsy — both diagnoses should be considered; do not assume all events are PNES simply because one captured event was nonepileptic
- Communication: Delivery of the diagnosis is therapeutic — frame as a recognized neurological condition, not "faking"; avoid the term "pseudoseizures"; use "functional seizures" or "nonepileptic events"
- Treatment: CBT is the evidence-based treatment (NES Treatment Trial showed significant reduction in seizure frequency); SSRIs for comorbid depression/anxiety; gradual ASM withdrawal when epilepsy has been excluded
- Emergency department: Avoid IV benzodiazepines for recognized PNES; unnecessary intubation has been reported; educate ED staff
- Prognosis: With appropriate diagnosis and treatment, approximately 50% of patients achieve significant reduction in episodes within 1 year; poor prognostic factors include longer duration of misdiagnosis, comorbid personality disorder, and ongoing litigation
Impact of Psychiatric Comorbidities on Epilepsy Treatment Outcomes
Psychiatric comorbidities are not merely bystander conditions in epilepsy — they directly affect epilepsy treatment outcomes in multiple ways. Depression is associated with a 2–3-fold increased risk of ASM intolerance and treatment failure. Patients with comorbid depression or anxiety are significantly less likely to adhere to their ASM regimen. Psychiatric comorbidities predict poorer surgical outcomes: patients with preoperative depression or psychosis have lower rates of seizure freedom after epilepsy surgery. Depression is also an independent risk factor for SUDEP, potentially through mechanisms involving autonomic dysregulation and nocturnal seizure vulnerability.
The economic burden of psychiatric comorbidities in epilepsy is substantial. Patients with comorbid depression and epilepsy have significantly higher health care utilization (more hospitalizations, ED visits, and specialist consultations) and higher total health care costs compared with epilepsy patients without psychiatric comorbidity. These findings underscore that treating psychiatric comorbidities is not ancillary to epilepsy care but rather a core component of comprehensive epilepsy management that improves both seizure-related and non-seizure-related outcomes.
References
- Keezer MR, Sisodiya SM, Sander JW. Comorbidities of epilepsy: current concepts and future perspectives. Lancet Neurol. 2016;15(1):106–115.
- Josephson CB, Jette N. Psychiatric comorbidities in epilepsy. Int Rev Psychiatry. 2017;29(5):409–424.
- Mula M, Brodie MJ, de Toffol B, et al. ILAE clinical practice recommendations for the medical treatment of depression in adults with epilepsy. Epilepsia. 2022;63(2):316–334.
- Mula M, Kanner AM, Jette N, Sander JW. Psychiatric comorbidities in people with epilepsy. Neurol Clin Pract. 2021;11(2):e112–e120.
- Tellez-Zenteno JF, Patten SB, Jette N, Williams J, Wiebe S. Psychiatric comorbidity in epilepsy: a population-based analysis. Epilepsia. 2007;48(12):2336–2344.
- Pham T, Sauro KM, Patten SB, et al. The prevalence of anxiety and associated factors in persons with epilepsy. Epilepsia. 2017;58(8):e107–e110.
- Klein P, Devinsky O, French J, et al. Suicidality risk of newer antiseizure medications: a meta-analysis. JAMA Neurol. 2021;78(9):1118–1127.
- Mula M, McGonigal A, Micoulaud-Franchi JA, et al. Validation of rapid suicidality screening in epilepsy using the NDDI-E. Epilepsia. 2016;57(6):949–955.
- Kim DH, Kim YS, Yang TW, Kwon OY. Optimal cutoff score of the NDDI-E for detecting major depressive disorder: a meta-analysis. Epilepsy Behav. 2019;92:61–70.
- Agrawal N, Mula M. Treatment of psychoses in patients with epilepsy: an update. Ther Adv Psychopharmacol. 2019;9:2045125319862968.
- Kanner AM, Schachter SC, Barry JJ, et al. Depression and epilepsy: epidemiologic and neurobiologic perspectives. Epilepsy Behav. 2012;24(2):156–168.
- Nadkarni S, Devinsky O. Psychotropic effects of antiepileptic drugs. Epilepsy Curr. 2005;5(5):176–181.
- Rugg-Gunn F. Adverse effects and safety profile of perampanel: a review of pooled data. Epilepsia. 2014;55 Suppl 1:13–15.
- Scott AJ, Sharpe L, Thayer Z, et al. Design and validation of two measures to detect anxiety disorders in epilepsy: the EASI and brEASI. Epilepsia. 2019;60(10):2068–2077.
- Micoulaud-Franchi JA, Lagarde S, Barkate G, et al. Rapid detection of generalized anxiety disorder and major depression in epilepsy: validation of the GAD-7. Epilepsy Behav. 2016;57(Pt A):211–216.
- LaFrance WC, Kanner AM, Hermann B. Psychiatric comorbidities in epilepsy. Int Rev Neurobiol. 2008;83:347–383.
- Salzberg M, Kumar G, Supit L, et al. Early postnatal stress confers enduring vulnerability to limbic epileptogenesis. Epilepsia. 2007;48(11):2079–2085.
- Keezer MR. Epilepsy comorbidities. Continuum (Minneap Minn). 2025;31(1):232–246.
- Clancy M, Clarke M, Connor DJ, Cannon M, Cotter D. The prevalence of psychosis in epilepsy: a systematic review and meta-analysis. BMC Psychiatry. 2014;14:75.
- Tian N, Cui W, Zack M, et al. Suicide among people with epilepsy: a population-based analysis. Epilepsy Behav. 2016;61:210–217.